Topics

Noisy RX on ubitx v5 #ubitx-help #filters #v5 #noise #bfo


George Metaxas
 

Hi there! I have built ubitx v5 inside VU3SUA's enclosure. With a dipole antenna for 40 m with ATU, I have been trying to receive signals. Things seem OK for CW, but SSB signals come in with reasonable noise that renders them unreadable. However, hooking up the same antenna on an RTL SDR with downconverter and tuning on the same frequency, SSB signals come in very nice.What should I do from the following?
1. Insert RF chokes / ferrite beads in the power supply lines of the IF amplifier stages, in an effort to reduce the noise that enters those stages.
2. Squeeze / spread the turns on the RF impedance transformers before and after the 45 MHz crystal filter, in case some mismatch exists.
3. Place a copper foil shield around the 45 MHz crystal filter and the matching transformers to improve the filter performance, especially the stop and rejection.
4. Replace the 2-pole crystal filter with a 4-pole crystal filter of the same package, bandwidth and impedance.
5. Try some power supply filtering on the audio stages before the LM386 amplifier (U1 on the ubitx PCB).

Any suggestions on how to proceed?

73, George, SV1DEQ 


 

Probably BFO needs calibration.

If you could make a recording and share with us, we will have a better idea of what you are facing.
For V5 add a series resistor of 1-10K in series with the volume control center lead. It wont affect
your volume but cure any instability on the audio chip.

Raj

At 04/10/2020, you wrote:
Hi there! I have built ubitx v5 inside VU3SUA's enclosure. With a dipole antenna for 40 m with ATU, I have been trying to receive signals. Things seem OK for CW, but SSB signals come in with reasonable noise that renders them unreadable. However, hooking up the same antenna on an RTL SDR with downconverter and tuning on the same frequency, SSB signals come in very nice.What should I do from the following?
1. Insert RF chokes / ferrite beads in the power supply lines of the IF amplifier stages, in an effort to reduce the noise that enters those stages.
2. Squeeze / spread the turns on the RF impedance transformers before and after the 45 MHz crystal filter, in case some mismatch exists.
3. Place a copper foil shield around the 45 MHz crystal filter and the matching transformers to improve the filter performance, especially the stop and rejection.
4. Replace the 2-pole crystal filter with a 4-pole crystal filter of the same package, bandwidth and impedance.
5. Try some power supply filtering on the audio stages before the LM386 amplifier (U1 on the ubitx PCB).

Any suggestions on how to proceed?

73, George, SV1DEQ


George Metaxas
 

Good idea Raj, I will do it as soon as I can and report it here. And I will try that resistor in series with the input signal, too. However, loud AM stations are received OK on ubitx, therefore that's why I do not touch the BFO for the time being. 


Evan Hand
 

George,

I had similar issues and added a "hiss" reduction to the LM386.  In essence that is a "Bass Boost" or low pass filter to allow for higher volume settings without the annoying static.  I got that from the LM386 datasheet.  Look at figure 18 on the sheet from TI:
https://www.ti.com/lit/ds/symlink/lm386.pdf

On one other note, I found that I had issues with the external audio amp used with the LM386 that can come with the kit from Sunil.  In the LM386 datasheet, it explains how to reduce the gain.  I did that to reduce feedback from the external amp.  Since I cut the gain of that section, it also cut the low-frequency boost to some extent.  At some point, I may go back and experiment.

I did the mods to the LM386 chip directly, bending the now unconnected pins out so that they were not connected when plugged in.  I also soldered the RC filter direct to the IC.  I have many spares of the IC, but only one v5 uBITX board.  If I mess up in the construction, I only lose a lower cost IC.

FWIW,
73
Evan
AC9TU


Evan Hand
 

I opened my v5 case to take a picture of how I added the bass boost to the LM386.

 


Jim Tibbits
 

Futzing around with my V3, I used this to alter bandpass of lm386 (located in ab7vf folder in photos section)



Gerard
 

Evan, i have put a message for you, in the post of the quality tricoid


George Metaxas
 

Thanks to all of you for your answers on that; following Raj's suggestion, I just recorded what I can hear on both receivers, the uBITX and the RTL-SDR on 40 m. I use exactly the same ATU settings on both cases.
Here is what the RTL-SDR hears at 7.130 MHz (SV1DEQ_RTLSDR_7130_000kHz-4.mp3)
and here is what the uBITX hears at 7.1273 MHz (recorded using my smartphone placed over one part of the headphones) (SV1DEQ_ubitx_7127300kHz.mp3)

It would be nice of I had a 3-dB power splitter so that I could feed both receivers simultaneously. Then, I could listen to both radios at the same time.
Anyway - it might be that either master calibration, BFO calibration or both are slightly off (some 10s of Hz perhaps). However, I must note that I have checked the radio with a signal generator and the frequency indication is accurate down to a few Hz.
Another point has to do with what Evan suggests about the LM386 amplifier; I think that I have lots of "pop-corn style" noise. It seems a good way to go to add that hiss reduction circuit branch around the LM386 amplifier.

Let me know of what you think after listening to the samples.


Evan Hand
 

George,

To me, it sounds like you do not have the signal tuned in.  The other alternative is that the BFO needs to be adjusted.  Before making the LM386 mods, I would recheck the BFO tuning.  Remember to write down the setting, as every time you go into the setup menu it is lost.

You can use this video and BFO tuning tool to help with the setup. 
https://www.youtube.com/watch?v=t6LGXhS4_O8

The calibration part will not work with the stock v5 software.  That software goes into transmit mode and you are supposed to use a frequency counter or another receiver close by the connected dummy load to measure or zero beat on the other receiver. 

NOTE:  Always have a dummy load connected when adjusting the stock v5 calibrations.  It will put out a 10MHz signal when you go into the calibration setting.  This is not in any of the ham bands, so you do not want to connect to a 10MHz antenna.

I would wait for Raj or others to weigh in before you do anything.

73
Evan
AC9TU


 

First set the BFO to 11.0555 (MHz) and try again. SSB may become OK.

You have quite a bit of noise in your location.

Raj

At 05/10/2020, you wrote:
Thanks to all of you for your answers on that; following Raj's suggestion, I just recorded what I can hear on both receivers, the uBITX and the RTL-SDR on 40 m. I use exactly the same ATU settings on both cases.
Here is what the RTL-SDR hears at 7.130 MHz (SV1DEQ_RTLSDR_7130_000kHz-4.mp3)
and here is what the uBITX hears at 7.1273 MHz (recorded using my smartphone placed over one part of the headphones) (SV1DEQ_ubitx_7127300kHz.mp3)

It would be nice of I had a 3-dB power splitter so that I could feed both receivers simultaneously. Then, I could listen to both radios at the same time.
Anyway - it might be that either master calibration, BFO calibration or both are slightly off (some 10s of Hz perhaps). However, I must note that I have checked the radio with a signal generator and the frequency indication is accurate down to a few Hz.
Another point has to do with what Evan suggests about the LM386 amplifier; I think that I have lots of "pop-corn style" noise. It seems a good way to go to add that hiss reduction circuit branch around the LM386 amplifier.

Let me know of what you think after listening to the samples.


George Metaxas
 

Hi again, it seems that I managed to make my uBitx sound better.
Raj, I set the BFO to 11.05555 MHz and I confirmed that I could listen to incoming stations. You were right!
Evan, I tried the bass boost circuit and the hiss is greatly suppressed. I soldered a 10 kOhm resistor in series with a 10 nF ceramic capacitor  across pins 1 and 5. Now audio is more enjoyable.
Here is a sound clip of how my radio hears. You can compare that to the ones posted previously.
Again, thanks to all of you for your help.
73, George, SV1DEQ


Viktors Miske
 

Had same issue here. Painstaking BFO adjustment in need. And heed the advice above writing down settings before saving. Twisted my lone brain cell here until Evan told me settings go away everytime you click on option.
Running a V5 here, no mods needed! Sweet rig out of the box to say the least. 

On Wed, Oct 7, 2020, 8:07 AM George Metaxas via groups.io <gm_rann=yahoo.gr@groups.io> wrote:
Hi again, it seems that I managed to make my uBitx sound better.
Raj, I set the BFO to 11.05555 MHz and I confirmed that I could listen to incoming stations. You were right!
Evan, I tried the bass boost circuit and the hiss is greatly suppressed. I soldered a 10 kOhm resistor in series with a 10 nF ceramic capacitor  across pins 1 and 5. Now audio is more enjoyable.
Here is a sound clip of how my radio hears. You can compare that to the ones posted previously.
Again, thanks to all of you for your help.
73, George, SV1DEQ


Evan Hand
 

George, 
That does sound good!  Glad you got it solved, now have fun!

73
Evan
AC9TU


_Dave_ AD0B
 

I bought a ubitx that was described as being deaf. On mine, adjusting the receive frequency alters the bfo. 
--
73
Dave
ADOB
Raduino bracket and Ham_Made_Keys


George Metaxas
 

Hi again, I have a small update on the excessive noise that you may have noticed in the audio samples. The reason for that is that my speaker is connected to the PAM8610 amplifier board using two-wire shielded audio cable and the shield is grounded on the aluminum case only on one end, close to the speaker, the other one left floating. I connected the shield of that cable to the case, close to PAM8610, and now the noise has been suppressed even more. KD8CEC firmware displays S5-6 after that mod;before that, I had noticed that noise resulted to S9 or even more, masking audio from stations. 
BTW, although slightly off-topic, are there any suggestions for a common-mode filter inside the ubitx case? 


Jonas Sanamon
 

Hi George,

I would suggest looking at the S-meter while removing power to the PAM8610 amplifier module. 
I've also been playing with using a Class D audio amplifier in my V.4 uBitX, but so far I've not been happy with it. I also tried to install it directly onto the speaker for minimum speaker cable length, and it got better but it seems the amp+speaker is too close to the uBitx board because I can hear the hiss increase as I put the lid in place.
Thinking of using LM386 instead now. I guess signals with square edges should be kept away from HF receivers :-)

Regards, Jonas


Den fre 16 okt. 2020 kl 00:56 skrev George Metaxas via groups.io <gm_rann=yahoo.gr@groups.io>:

Hi again, I have a small update on the excessive noise that you may have noticed in the audio samples. The reason for that is that my speaker is connected to the PAM8610 amplifier board using two-wire shielded audio cable and the shield is grounded on the aluminum case only on one end, close to the speaker, the other one left floating. I connected the shield of that cable to the case, close to PAM8610, and now the noise has been suppressed even more. KD8CEC firmware displays S5-6 after that mod;before that, I had noticed that noise resulted to S9 or even more, masking audio from stations. 
BTW, although slightly off-topic, are there any suggestions for a common-mode filter inside the ubitx case? 
_